Multimodal Word Sense Disambiguation in Creative Practice

Guevara, Manuel Ladron de; George, C.; Gupta, Abhinav; Byrne, Daragh; Krishnamurti, Ramesh

doi:10.1109/icmla51294.2020.00055

Cited by 1 publication

(2 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Preliminary explorations that visually disambiguate vague terms in the context of design have been done by (Ladron de Guevara et al, 2020). The authors use a multimodal approach that combines a pretrained convolutional neural network to get the representation for images with general word indexes into a common joint subspace.…”

Section: Prior Work On the Coded Datasetmentioning

confidence: 99%

“…To address the issue of disentangling design intents in the context of creative practice, we use the CODED dataset, first presented in (Ladron de Guevara et al, 2020). The self-annotated CODED dataset contains a total of 33,230 samples of contemporary creative works represented by 264,028 raw sentences-provided by the original creators and by art curators-that describe 241,982 images.…”

Section: Dataset and Input Modalitiesmentioning

confidence: 99%

See 1 more Smart Citation

Proceedings of the 27th Conference on Computer Aided Architectural Design Research in Asia (CAADRIA) [Volume 1]

2022

CAADRIA Proceedings

View full text Add to dashboard Cite

The work collected in these volumes has been produced during extraordinary circumstances, as the Covid-19 pandemic has continued to impact working and living conditions. Papers have been written, reviewed and edited throughout lockdowns, challenging family situations and health crises. The high number of submissions and quality of the papers is a testament to the resilience and strength of the CAADRIA community.The papers in this publication have been selected through a two-stage, doubleblind peer review process. All reviews and papers have been evaluated by the Paper Selection Committee. After receiving an initial 488 abstract submissions, a final set of 150 full papers has been selected for publication. We thank all authors for their contributions to the process and congratulate the accepted paper authors.The papers in these proceedings are specifically selected for their contribution to this year's conference theme, following the conference organisers' call for authors to position their work in relation to the United Nations Sustainable Development Goals (SDGs). As the world is experiencing the increasing impacts of climate change, there is an urgent need to reflect on the potential of the latest research in architecture, urbanism and construction to address these global challenges.Our special thanks go out to our more than 180 international reviewers, who volunteered their valuable time and effort. We also thank the organising team in Sydney for developing the 2022 CAADRIA Conference as a strong collaborative effort between academic institutions, industry and practice. We thank Gabriel Wurzer for his support during the proceedings' final production stage, and for facilitating their publication via the open access database CumInCAD.org. Finally, we sincerely thank the CAADRIA community for offering us the honour to serve as members of the paper selection committee for Post-Carbon, the 27 th International CAADRIA Conference 2022.

show abstract

Section: Prior Work On the Coded Datasetmentioning

confidence: 99%

Section: Dataset and Input Modalitiesmentioning

confidence: 99%